More than half of all television shows contain sexual content, and rarely do these shows make reference to sexual responsibility issues such as pregnancy or STDs. There is widespread concern about the effects of such portrayals on adolescents: at its heart is the worry that watching television may cause adolescents to engage in intercourse or other sexual activities earlier than they would otherwise. Yet little scientific research has been done to determine whether this is true. The proposed study will investigate the associations between viewing sexual content on television and sexual attitudes and behaviors in a national sample of 2000 adolescents between the ages of 12 and 17. The study extends previous work by examining this question longitudinally while controlling for other predictors of sexual activity, thereby laying the groundwork for making causal inferences. The core of the study is a national survey, with two waves of data collection one year apart. Data concerning television viewing, sexual attitudes and activities and background factors will be collected. These data will be combined with the findings from a separately funded state-of-the-art content analysis covering the television season intervening between these two surveys. This unique opportunity will allow us to develop highly sensitive measures of exposure to diverse forms of content, including content portraying risks and responsibilities that may be educational. Thus, we will be able to provide a state-of-the-art answer to the questions concerning both the potential positive and negative effects of television sexual content on adolescent viewers.